The broad long-term objectives of this project are to determine the biological functions of secreted phospho-protein 1 (Sppl, also known as osteopontin or 2ar). Elevated expression of this protein closely correlates with tumorigenicity of rodent and human cells, and Sppl expression is induced in mouse epidermis with tumor promoters. Identification of high levels of Sppl in the serum of patients with disseminated carcinomas provides evidence for a strong correlation between Sppl expression and progressive tumor growth, and, moreover, provides direct evidence for an association between elevated Sppl expression and human cancer. Several findings suggest the hypothesis that Sppl promotes tumor cell invasion; in particular, Sppl interferes with cell attachment to extracellular matrix (ECM) and induces expression of the matrix degrading protease, stromelysin (in zymogen form). The proposed experimental plan will elucidate further the effects that Sppl exerts on tumor and host cells that may facilitate tumor cell invasion and will determine the mechanisms by which Sppl exerts these effects. The Specific Aims are: (1) determine if Sppl binds to ECM components, if it prevents attachment to ECM components in addition to vitronectin, and if it promotes tumor cell penetration of ECM, (2) define the inductive effects of Sppl (native and thrombin-cleaved) on synthesis of secreted proteases (especially stromelysin) by normal and tumor cells, (3) characterize the cell surface receptors (presumably inte- grins) which bind native and thrombin-cleaved Sppl, and (4) define the distribution of Sppl and Sppl mRNA in human tumors with immunohistochemistry and in situ hybridization and compare with the distribution of stromelysin and stromelysin mRNA. Binding of Sppl to ECM components and effects of Sppl on cell attachment to ECM will be addressed with protein binding and cell attachment assays, respectively. Effects of Sppl on tumor cell invasion will be assessed in vitro by quantitating tumor cell penetration (in the presence and absence of anti-Sppl antibodies) of ECM coated filters and amniotic membrane. The inductive effects of thrombin-cleaved Sppl on normal and tumor cell protease expression will be determined following addition of Sppl and/or anti-Sppl antibodies. Expression of stromelysin and collagenases will be quantitated with several techniques (immunoprecipitation, Northern blotting, substrate assays). Cell surface receptors (presumably integrins) for Sppl and thrombin-cleaved Sppl will be identified with affinity chromatography and monoclonal antibodies. These experiments will determine whether Sppl and thrombin-cleaved Sppl mediate their different effects via the same or different receptors and if they bind to vitronectin receptors. Finally, detailed immunohistochemical studies on the distribution of Sppl in human benign and malignant tumors, and in comparison with the distribution of stromelysin, will be performed to determine the relationship of these two proteins to tumor cell invasion in vivo. In situ hybridization with nucleic acid probes will provide for precise localization of expression of these two proteins.